Summary

This tool can be used to ingest data into a cluster via RabbitMQ.

Prerequisites

RabbitMQ
Python >3.6
PSQL on multiple cluster instances
Partitioned schema on the nodes such, that the partition key is hashed with a custom modulo function

Have your partitions such, that the partition key can be used as routing key. For instance, on 2 datanodes with 4 partitions in total, ensure, that all data hitting node #1 with key modulo 4 partitions goes to partitions 0 and 1, whereas on node #2 it would go to partitions 2 and 3.
On each node, start a receiver, e.g. assume you have 2 nodes, then you would start one receiver on each:

./receiver.py --rmq-dsn --receiver-id 0 --rpc-exchange myrpc --dsn ./receiver.py --rmq-dsn --receiver-id 1 --rpc-exchange myrpc --dsn

Note the different receiver IDs.
Start the ingestion on the sender node with two receiving data nodes and for data coming from a textfile one would do:

cat mydata | ./sender.py --rmq-dsn --receivers 2
--exchange myexchange --rpc-exchange myrpc
--table --db
--data-source stdin